2024-12-13 19:29:10 RT @jonathanstray: Newsrooms could regain conservative audiences in 2025. There are things they could try, and publications doing this succ…

2024-12-13 19:28:59 I wrote about how "mainstream" media has made a similar mistake in covering the UnitedHealthcare CEO murder as in covering Trump, which is part of the reason it's becoming less mainstream to a lot of America https://t.co/sSzGmz5jUR https://t.co/vzvuRQ2eMl

2024-12-13 19:26:41 RT @DevadasRajaram: “We cannot be neutral about this, by definition. A free press that doesn’t agitate for democracy is an oxymoron.” - @gl…

2024-11-11 15:26:14 @dkeaneactual And I’m not advocating banning bikes from the roads. I’m saying that could happen if autonomous cars become much more widespread.

2024-11-11 15:25:13 @dkeaneactual I don’t think access to credit is the reason. As the piece says, off-road bike sales have surged. Also, though it isn’t in the piece, truck sales have also exploded over that period. People have money. They just aren’t choosing to spend it on (street) bikes.

2024-11-10 18:42:31 I wrote a thing about motorcycling across the U.S., and motorcyclists hate it https://t.co/kJI1WhuZo1

2024-11-09 08:19:26 @AnandWrites Read @TaylorLorenz on why this is difficult: among other reasons, positions of progressive influencers are at odds with the people who would fund them https://t.co/dOYpMOFwRi

2024-11-08 14:36:05 This is an excellent analysis. And if you wrote it at 1am even more so https://t.co/PfLbIdWyio

2024-11-06 13:50:43 Call or message your trans friends today. They are more scared than just about anybody.

2024-10-25 13:42:45 The polls are still neck-and-neck, but if you're starting to feel dread at the possibility that Trump may win, here are some aspects of democracy that I think will flourish even under his presidency. https://t.co/q6OBMMlYai https://t.co/8gjgnmYEYF

2024-10-13 17:52:36 RT @mitpress: Cambridge Forum continues its series AI: Servant or Master with @gregmepstein exploring the global worship of technology and…

2024-10-04 18:29:37 RT @glenweyl: Congratulations to @glichfield on the first English language piece on the most exciting thing I think happened this year in g…

2024-10-04 16:54:01 What if each voter could talk directly to an AI version of a candidate for office, grill them about their policy platform, and make suggestions for improving it? That's what a young political outsider tried in Tokyo's gubernatorial election this summer. https://t.co/dgoZPAcB0z https://t.co/y3CSX4rWN1

2024-10-02 18:42:19 Yesterday’s VP debate was a lot more policy-focused than the Trump-Harris faceoff. All the more reason, I’d argue, to use real-time AI-enabled fact-checking to pick out candidates’ lies and exaggerations. https://t.co/iCEMulPfOk https://t.co/7O6XlQhLOI

2024-09-27 14:57:31 Big Tech has huge power over government, since it provides much of government's critical infrastructure. In this week's FUTUREPOLIS I talk to Marietje Schaake, author of the new book "The Tech Coup," about what governments can do to seize back control. https://t.co/PiExDt4QBz https://t.co/gL3a4eNFwk

2024-09-24 18:56:11 I'm speaking at @thesfcommons on Oct. 14 about AI &

2024-09-20 21:00:27 RT @pulitzercenter: A special thank you to @macfound for organizing this engaging #ONA24 session on elevating AI accountability coverage in…

2024-09-20 21:00:03 RT @pulitzercenter: If you are an editor increasingly tasked with commissioning &

2024-09-20 19:04:03 My newest issue of FUTUREPOLIS looks at an experiment in how to talk about the most polarizing topic in America: guns. Eleven strangers ranging from a firearms instructor to a schoolteacher who has buried nine of her students came together. https://t.co/DONVziABnR https://t.co/YWuk2hZotK

2024-09-13 15:24:39 Last week I launched FUTUREPOLIS, a newsletter on reinventing democracy for the 21st century. For today's issue, I wondered: What if ABC had used AI to fact-check the presidential debate in real-time? So I ran an experiment to find out. https://t.co/csbWUvGQxX https://t.co/QwqKbmvCaW

2024-09-11 01:21:48 RT @MarinaWalkerG: Attention editors increasingly tasked with helping shape coverage of AI technologies, we have you covered The AI Spotl…

2024-08-20 16:34:54 I'm excited to be one of @UCBerkeley's #TechPolicy Fellows for 2024-25! https://t.co/KHgGFSG8XP https://t.co/9X3SuK6Hdy

2024-08-08 16:23:06 @halophoenix @WIRED @PCMag Congrats, Alan! Enjoy the new gig.

2024-07-18 22:39:05 Nothing says “Millennial VP candidate” more than this https://t.co/ENvgmCxQJd

2024-07-08 16:31:39 And does the city have a plan for disposing of the millions of non-wheelie bins that New Yorkers currently put trash into? https://t.co/q6RAnw6tfZ

2024-06-28 02:01:46 Watching this debate I have to keep reminding myself that the whole reason I became a U.S. citizen was in case Trump wins.

2024-06-28 01:58:07 @ftrain @sally_j I hope you’re doing your breathing exercises right now

2024-06-19 00:55:36 @SarahScire @BusinessInsider @OpenAI @decka227 @InsiderUnion Do an interview with an expert on LLMs on this. My guess is the answer would be that it’s impossible to prevent hallucinations but might be possible to bolt on some external software for finding URLs that match the content of the LLM’s answers - basically a search plugin for LLMs

2024-05-27 11:58:50 It’s a serious flaw in the ride-hailing model that as a rider you can penalize the driver but not the platform (for eg consistently underestimating wait times). Total lack of accountability and not enough competition to make switching a meaningful option.

2024-05-04 13:45:37 RT @MarinaWalkerG: Thank you @decka227 @NiemanLab for featuring the AI Spotlight Series &

2024-03-01 00:00:00 CAFIAC FIX

2024-03-11 00:00:00 CAFIAC FIX

2023-05-22 19:33:27 RT @KimZetter: Australian guy plagiarized SolarWinds story I wrote for @WIRED by creating ad-supported @Youtube vid in which audio is just…

2023-05-22 16:14:00 @jonathanstray @metaphdor I'm about 25% into and enjoying Stephen Markley's "The Deluge" but it's possibly too close to reality to be called sci-fi

2023-05-19 19:00:00 CAFIAC FIX

2023-05-21 19:00:00 CAFIAC FIX

2023-05-03 19:54:10 @LaurenGoode @LidianeJones That was actually me in the giant cat costume

2023-05-03 15:50:52 RT @LaurenGoode: We sat down with @LidianeJones, Slack’s new CEO, to talk about the future of hybrid work, automation, and whether tech can…

2023-05-02 16:41:11 Two years after the biggest ever supply-chain attack came to light, there are still unanswered questions about how it was launched and how deep it went. @KimZetter tells the blow-by-blow of how the SolarWinds hack was discovered and what we know so far. https://t.co/0UK2aDcOvg https://t.co/xg3a9D0BXa

2023-05-01 21:14:50 Just in case we never get logged back in again I’m glichfield at bsky, T2 and on mastodon dot world

2023-05-01 17:18:38 What's in Elon's mystery magnet? by @GregoryJBarber https://t.co/ZGAWxEf2Jy

2023-04-26 14:59:59 “Studying these extremely long-lived people is like studying Usain Bolt when it comes to running and saying, ‘Yeah, we can all run that fast,’” @sjayolshansky tells @mattsreynolds1 in the first in our @WIRED series on life extension. https://t.co/lkPPsOCWKY

2023-04-26 14:02:42 Did I mention that @LaurenGoode and I have a new podcast? This week @nraford explains why he spotted Covid early and why you should prepare your kids to become climate refugees https://t.co/ogzxxeZ4MY

2023-04-21 06:36:41 @runasand @InfoSecErnest @stevanzetti @WIRED @MattWalshBlog @JoeUchill Once again - Twitter’s policy does not ban soliciting hacked materials, only making them public. And that screenshot had already been published by Walsh himself when Dell posted it. Countless others are now spreading it, and not having their accounts suspended.

2023-04-21 00:00:01 CAFIAC FIX

2023-04-12 14:21:17 RT @LaurenGoode: Our brand new @WIRED podcast ~Have a Nice Future~ has launched! And in our first episode, @glichfield and I took a trip to…

2023-04-06 16:35:09 @charlesarthur @jason_pontin That works too

2023-04-06 06:05:07 @charlesarthur @jason_pontin If a single word distracts you that much may I suggest meditation techniques

2023-04-05 14:21:30 RT @epsneider: “Platform.” “Database.” “Swipe.” How did these words get lodged in our collective vocabulary, and what do they conjure? 1/3

2023-04-05 14:19:36 @katecrawford @lexfridman @_KarenHao I had a similar experience when I tried using it for story ideas. Every single URL here is made up. It's hard to get people to grasp that these models don't encode information about the world, only information about the words that people use about the world https://t.co/T89MZbfAYJ

2023-04-05 14:13:27 A little news... @LaurenGoode and I are launching a podcast! HAVE A NICE FUTURE will be a weekly chat with someone shaping the weird, exhilarating, scary future we're racing into, asking: is this the future we want? And if not, how do we get a better one? https://t.co/EnaJFCKkzT

2023-04-05 00:42:32 @charlesarthur @jason_pontin Nobody said this was the ~primary~ meaning. And thank you, I know how to use “disinterested” to mean “unbiased” and I’m sure Justin Smith does too. I don’t think anybody could mistake his intention in the way he used it.

2023-04-04 21:48:57 @jason_pontin But also, from the OED: https://t.co/eUBequIOzY

2023-04-04 21:15:25 Requirements for Amazon drone delivery: - no overhanging trees or power lines - 10-foot radius of clear ground - stay out of the backyard during delivery window - only Monday thru Thursday - no rain or strong wind - Amazon observers within line of sight https://t.co/SjyV0JZLBA

2023-04-04 19:14:53 @jason_pontin Just you try arguing with a philosopher about the meaning of words

2023-04-03 18:26:00 "I called a few people... to learn more about their positions—and ended up on the phone for hours, feeling for all the world like a kid listening to her beloved but divorcing parents bitterly complaining." @Knibbs on the internet archive lawsuit https://t.co/zZdUc7rt9q

2023-03-24 16:51:46 @Kantrowitz In the future everyone will have a podcast for 15 minutes

2023-03-24 16:49:40 @andyparsons @stevenerat @CNN @ContentAuth @C2PA_org Thanks for the correction. Am I right in thinking that the only way to create CAI metadata right now is with Adobe products, or are there other platforms that already enable this?

2023-03-24 15:05:21 @acroll Generative Adversarial Generative ARtificial Intelligence Networks (GAGARINs)

2023-03-24 04:02:39 @stevenerat @CNN @ContentAuth Content Authorization is a good idea but depends on all creators using Adobe products and also on platforms that publish content using the same system to preemptively check and flag content that may be faked. Otherwise, you only find out it’s fake if you go digging.

2023-03-24 02:41:12 Going on @CNN at 11pm ET to talk AI/Trump deepfakes, TikTok hearings, and more

2023-03-23 23:23:08 Not having a blue checkmark is going to be the new blue checkmark https://t.co/8J7P47wXgV

2023-03-20 17:31:05 RT @WIRED: Join us tomorrow at #WIREDHealth to explore and map the future of health. The event gathers medical practitioners, health techno…

2023-03-20 17:30:52 RT @GregWilliams718: So excited that @WIRED Health is returning next week, with a stellar line-up including: @FryRsquared @anilkseth @nicol…

2023-03-20 13:53:27 @ftrain I'm so sorry for your loss, Paul. James Baldwin's collection "Jimmy's Blues" has a lot of wry humor, including my current favorite, "Inventory/On Being 52" https://t.co/xxtCvJiKC2

2023-03-20 13:37:37 Una gran pena recibir la noticia de la muerte de Jean-François Fogel, maestro y mentor para generaciones de periodistas de varios países, incluyéndome a mí. https://t.co/6RFo4lADiN

2023-03-11 23:37:17 If you're at #SXSW2023 and you're affected by the #SiliconValleyBank collapse, DM me or come find me. I want to hear your story.

2023-03-11 23:35:38 @steampunk_explr @SarahScire @WIRED I had not seen your policy, but yes, the thinking is very similar!

2023-03-11 01:24:47 RT @DrawAndCode: Tomorrow at #SXSW2023, our creative extraordinaire @Coopalskii will take the #UKHouse stage with @itsyush @dimensionstudio…

2023-03-10 21:31:31 "It's multimodal" is going to be the new "it's got electrolytes"

2023-03-09 22:55:46 Our second @SABEW winner was @laurensmiley's deeply nuanced profile of Rafaela Vasquez, the operator of the first self-driving car to kill a pedestrian—a landmark case study in a question for the ages, of how to assess liability for autonomous systems https://t.co/2nKM4v3cLL

2023-03-09 22:51:09 (That story is from Andy's book TRACERS IN THE DARK, which from which we also excerpted the account of the takedown of AlphaBay, an even bigger drug and crime website than the notorious Silk Road https://t.co/uk4XzqXQx9)

2023-03-09 22:47:54 One is @a_greenberg's nail-biting account of how an international coalition of law enforcement cracked and took down the world's biggest child sexual abuse site by tracing its users' supposedly anonymous cryptocurrency payments https://t.co/QkvsADEtt1

2023-03-09 22:43:15 Two wins for @WIRED in the @SABEW Best in Business Awards https://t.co/6s7LjsuHsv ...

2023-03-09 00:12:05 "Q. Do you still feel connected to the work you wrote decades ago? A. As when I was writing it? Of course not. Q. Do you even feel like the same person? A. No. I got older. It would be bizarre if I was the same person. I’d be a vampire." Pure Atwood https://t.co/2omiYXZ8jL

2023-03-08 23:34:22 https://t.co/o99f6pJD7X

2023-03-08 23:33:55 https://t.co/ntE2Zb81PI

2023-03-08 23:32:48 @WIRED @mattsreynolds1 https://t.co/fTA0t2tLJQ

2023-03-08 23:32:23 Congrats to @WIRED's @mattsreynolds1 for being named the UK Press Awards' science &

2023-03-07 19:32:06 Here's part 2... https://t.co/BuY4orN7Dk

2023-03-07 19:31:03 ... and for support from @pulitzercenter, @eyebeamnyc's Center for the Future of Journalism, and our expert reviewers @mmitchell_ai @achould @informusiccs @jannspiess Chris Snijders @nicolaskb @ndiakopoulos and Nripsuta Saxena

2023-03-07 19:29:45 @LHreports @gabriels_geiger @EvaConstantaras @jus_braun @HtetAung04 @EvalineSchot @soizicpenicaud @pjarandia ... and from @WIRED: @dmehro, Kyle Thomas, Alyssa Walker, @katherinlamart, Raagul Nagendran, Hari Moorthy, Ishita Tiwari, @DanieC @lilyelle @AndrewCouts @jtemperton @tsimonite ...

2023-03-07 19:21:22 Many thanks to the people who put so much time in from @LHreports: @gabriels_geiger @evaconstantaras @jus_braun @HtetAung04 @EvalineSchot @soizicpenicaud @pjarandia ...

2023-03-07 19:18:03 @LHreports @WIRED And if you're a journalist and want to know how to do this kind of thing in your own newsroom, join a free webinar hosted by @LHreports and @WIRED on March 9 at 4pm CET/3pm GMT/10am EST/7am PST https://t.co/Zh0BXm3zq9

2023-03-07 19:15:16 This series was a months-long collaboration between @LHreports and @WIRED. You can read more on the methodology here https://t.co/ytYdBNDl6S

2023-03-07 19:12:44 Read the entire series and try our simplified model of one such risk-scoring algorithm out for yourself here https://t.co/YHmqfFOWTe

2023-03-07 19:05:37 As a result, when the algorithm's dragnet snares innocent people, it can be next to impossible to understand why they were flagged, who to appeal to, or how to improve the system to reduce false positives in future. People can fight for years to clear their names and their debt.

2023-03-07 19:02:49 These firms, such as Accenture, Deloitte, Capgemini, and PwC, are not made accountable for the final results. They build black-box systems that the officials who operate them often poorly understand. And intellectual property agreements shield the systems from public scrutiny.

2023-03-07 18:57:21 The final story in the series takes a broader look at the algorithmic fraud-detection industry and shows the flaw at its heart: Lacking the expertise to build these systems in-house, governments typically subcontract them to consulting firms. https://t.co/eft1533EE4

2023-03-07 18:52:40 One Danish fraud investigator we interviewed says that the algorithmic system has not been more reliable than old-school methods like tips from schools and social workers. But many investigators rely on it more, and may come to trust it more because it's supposedly objective.

2023-03-07 18:48:34 The documents we obtained suggest Denmark's system uses variables that could be equated with ethnic profiling, which would likely make it fall foul of the EU's proposed AI act.

2023-03-07 18:46:21 In the first of them, we look at how conservative politicians' obsession with supposedly rampant benefits fraud led Denmark to adopt an intrusive algorithmic surveillance system, though only 8% of the cases it flagged have led to punishment https://t.co/mMrSId3Adm

2023-03-07 18:41:41 Today @WIRED runs the final two instalments in "Suspicion Machine," our joint investigation with @LHreports into how algorithms designed to detect welfare fraud end up snaring innocent people and perpetuating gender and racial bias. https://t.co/rEfNTJaTat

2023-03-07 18:37:26 Congratulations to @amitkatwala for his new book on the flawed origins of the lie detector, of which you can also read an excerpt in @WIRED (see his next tweet) https://t.co/J7cWzCrKYE

2023-03-06 18:03:41 RT @SarahScire: Readers deserve better than "plausible-looking." Talked to @glichfield about @WIRED's new generative AI policy, the first I…

2023-03-06 18:01:57 More stories coming this week in this investigative series, "The Suspicion Machine," from @WIRED and @LHreports - stay tuned. https://t.co/rEfNTJaTat

2023-03-06 18:00:35 The investigations can be incredibly stressful for people already in precarious circumstances. Some of them told us of losing benefits over minor mistakes, fighting years-long court battles to have benefits restored, and being driven to consider suicide. https://t.co/0fszq033nI

2023-03-06 17:55:27 Human investigators decide whether or not to pursue each case the algorithm flags. But we've seen in countless settings that over time, humans can come to trust a flawed algorithmic system and exercise less judgment of their own.

2023-03-06 17:51:53 Another odd feature of the algorithm was that it asked different questions about men and women—number of kids or recent addresses for women, for example, vs language proficiency or financial standing for men—making it hard to tell if it was assessing people fairly.

2023-03-06 17:43:50 The @LHreports / @WIRED investigation found that the system in Rotterdam was so blunt that if a caseworker made ANY notes in the file about someone's appearance, motivation etc, these had the same effect on the risk score whether the comments were positive or negative.

2023-03-06 17:40:40 (Twitter doesn't seem to be loading images rn, but you can see the interactive that shows how the risk scores are assessed at https://t.co/YHmqfFOWTe)

2023-03-06 17:38:48 For example, if a caseworker observes that a welfare recipient is not dressed for job interviews, seems dispirited, or struggles to deal with setbacks, that increases their risk score and makes them more likely to be flagged as a potential fraudster. https://t.co/y2Ht90EIwr

2023-03-06 17:25:04 @rachelmetz @technology Congratulations!

2023-03-06 17:24:10 Around the world cities have been building suspicion machines: Algorithms that flag potential welfare fraud. An investigation by @WIRED and @LHreports took one of these systems apart and showed how both algorithmic and human bias can skew its results. https://t.co/YHmqfFOWTe

2023-03-05 10:00:00 CAFIAC FIX

2023-03-02 22:00:00 CAFIAC FIX

2023-02-28 01:56:29 “Because everyone in the funeral industry is so invested in existing technologies, you need outsiders to help with thinking outside the box—no pun intended.” https://t.co/Np3kznLsXg

2023-02-28 01:54:23 Por si fuera poco la contaminación común y corriente en CDMX, ahora los microplásticos… por @annielagos https://t.co/SqaHYp2hsK

2023-02-27 01:00:00 CAFIAC FIX

2023-02-24 18:42:12 The @WIRED Headline Of The Week award goes to ⁦@dmehro⁩: “Amazon Has a Donkey Meat Problem” https://t.co/BKkfu3Ri9f

2023-02-23 20:30:12 @GiladEdelman Just wait, the next phase is wanting to write history books.

2023-02-23 20:27:43 RT @ASME1963: ASME Awards 2023: @WIRED wins Best Profile Photograph category for a photograph by @choutoo from “The Multifarious Multiplexi…

2023-02-21 18:14:56 RT @GregWilliams718: We're back on March 21st for our 11th @WIRED Health event in London with another stellar line-up covering reproductive…

2023-02-21 18:12:40 RT @epsneider: Hi! As of this month I’m editing for @wired Ideas, which means I want you to pitch me! Send your grand designs, hot takes, a…

2023-02-20 15:15:16 RT @1a: We're live, kicking off a new series with our friends over at @WIRED all about artificial intelligence. @glichfield, @SDobrin, and…

2023-02-20 14:59:53 About to go on @1a from @wamu885 and @NPR to talk about AI with @IreneSolaiman and @SDobrin - tune from 10-11am EST at https://t.co/7o6cpx82th

2023-02-14 22:59:15 And of course... https://t.co/00VUXCABdR

2023-02-14 22:58:47 https://t.co/ml464iYzxt

2023-02-14 22:58:30 https://t.co/lEHA2BexMd

2023-02-14 22:58:15 https://t.co/FpX6z2aWDt

2023-02-14 22:57:54 https://t.co/LFQZ97R9HZ

2023-02-14 22:57:32 Technology, sex and love? Hell yes! A thread of Valentine's Day-related stories https://t.co/oELMh3Egk3

2023-02-13 17:45:18 @bupbin @techreview I'm glad to see people at @techreview are still reading my style guide :) https://t.co/qqud8d2uXz

2023-02-10 18:50:05 Even if you're a Hogwarts Legacy fan, read @goodjaina's review and try to hear what it's saying. This gets deep inside some of the paradoxes of JK Rowling fandom, as well as making a key point about how much the games industry depends on queer creators. https://t.co/E8YIrJP8cp

2023-02-09 18:20:45 @jeffjarvis I wasn't gonna say anything, Jeff, but...

2023-02-09 16:18:19 @jjaron @jamesrbuk Sorry to quibble but neither Bard nor FT claimed JWST *discovered* a planet by imaging, only that it took the first image of one All of which just further demonstrates that AI is going to struggle to make sense of info online when we humans are so bad at doing it ourselves

2023-02-08 04:41:31 @superwuster @ChaldeanQueer Not an expert but I think you can’t accurately fly a balloon towards another balloon and there are no aircraft that can fly at 60k feet slowly enough to capture a slow-moving object like that

2023-02-07 21:35:38 Today feels like the day the generative AI arms race began, w/ Microsoft and Google both incorporating it into search and the latter adding chatGPT to its Edge browser, massively increasing adoption—by @AarianMarshall https://t.co/A3Lbu9y45i &

2023-02-07 02:28:02 @pwhall Previously in my career: cloud everything, mobile everything, apps for everything, HTML5 everything, social-media everything, AI everything, Uber-for-everything. The everything is everything. All else is transience.

2023-02-07 02:23:56 A story of how Russia became one of the most surveilled countries in the world, with an estimated 21 million surveillance cameras and virtually no legal constraints on the use of data for face recognition, by @MashaBorak https://t.co/hsCgdoREoe

2023-02-05 16:55:18 So maybe a 3-4x death rate among Chinese 90-somethings does indeed portend a much more severe surge? But there are so many other factors that could skew the numbers. I’d be curious what an epidemiologist makes of them.

2023-02-05 16:53:04 One interesting stat: in the first six months of the pandemic in the US, without vaccines, over-85s were dying at a 14.7% higher rate than pre-pandemic. That’s much lower than I thought. https://t.co/v8Y3XthyAT

2023-02-05 16:49:26 The article also doesn’t explicitly acknowledge what’s in the data it draws from: almost those obituaries are of people well over 90. So on the face of it, a 3-4x jump in their death rates during a Covid surge might not seem surprising. But how big has it been elsewhere?

2023-02-05 16:45:52 This story suggests that excess deaths among Chinese academicians point to a big undercount of Covid deaths by the Chinese govt. But it doesn’t address the most obvious question: how high an excess death rate would one *expect*? https://t.co/gwSC3RUPRV

2023-02-03 20:05:33 @rachelmetz Have you tried IMWT? https://t.co/xFrCNaEYH0

2023-01-30 01:00:00 CAFIAC FIX

2023-01-23 22:19:35 In which ⁦@doctorow⁩ explains his grand theory of why big tech platforms all get worse over time: their market logic means they inevitably end up abusing first the users and then the business customers on whom they relied for their success. https://t.co/aG17j7bhME

2023-01-22 21:04:42 RT @Kevin_Ashton: Are you on the For you tab right now when you don't want to be? If so, please consider retweeting this PSA to help a Twit…

2023-01-19 18:31:51 RT @a_greenberg: Pigs in the US are increasingly slaughtered with CO2 gas chamber systems that can asphyxiate as many as 1,600 pigs an hour…

2023-01-19 18:18:09 This is when you know you’ve lost the narrative https://t.co/YR55MTeasm https://t.co/X733KFubM6

2023-01-19 11:42:16 This discussion with the director of the FBI, the CEO of @Cloudflare, the first deputy PM of Ukraine, the CEO of @geckorobotics and me touched on an incredible array of topics under the umbrella of technology and its role in national security. https://t.co/U6Tcv5X4LA

2023-01-18 13:10:53 RT @PeterGuest: Now is as good at time as any to say I'm working @WIRED in London for a while, commissioning and editing some non-US storie…

2023-01-17 13:45:24 @jonathanstray ... and can do so because they have a following. i.e. maybe opinion change happens for tribal/loyalist reasons, and new information in media reporting is a necessary but not sufficient condition for that to happen.

2023-01-17 13:42:07 @jonathanstray Are these studies measuring direct causal impact? I'd hypothesize that in very rare cases a story changes opinion/behavior directly, but more often, it acts as ammunition/fodder for actors who want to influence opinion (e.g. politicians, celebs)...

2023-01-13 03:08:57 In which ⁦@amandahoovernj⁩ finds a silver lining in the gas stove culture wars: “More people are learning how their powerful gas stoves, perfect for searing steaks and heating up cast iron cookware, could adversely affect their health.” https://t.co/hDEL3K1Oav

2023-01-12 23:39:05 This is going to be me in 20 years https://t.co/Dmn3quOeLj https://t.co/IgONF8EFXB

2023-01-12 23:30:02 RT @epsneider: Hi! As of this month, I'm writing @WIRED Classics, a weekly newsletter highlighting stories from the archive each Saturday.…

2023-01-10 21:02:55 A belated welcome to @dellcam and @peard33, who started at @WIRED this week!

2023-01-09 19:33:15 "I’m a lubricant. I want people to feel comfortable. I don’t know how to function at the expense of anyone’s comfort level. I’m a people pleaser." Pedro Pascal explains to @hemjhaveri how he's completely unlike many of the people he plays on screen https://t.co/Zg2ws3Tz6q

2023-01-06 18:43:21 "In the 2020s the big shift is toward building with generative AI. This year thousands of startups will emerge with business plans based on tapping into the APIs of those systems." @StevenLevy on the next big tech gold rush. https://t.co/QepBjjsfJa

2023-01-05 13:57:34 A compelling case for a controversial idea: a regulated market for the sale of human organs, from @dylancwalsh https://t.co/hEVtbxhuf8

2023-01-03 22:44:55 RT @roseveleth: If last week you saw me Tweeting about the @flashforwardpod finale and were like "that seems cool but I don't want to liste…

2023-01-03 22:43:44 RT @emilylmullin: I'm looking to tell more stories about synthetic biology in 2023. Feel free to get in touch with pitches: emily_mullin@wi…

2023-01-03 19:14:21 Just as regulators used the threat of terrorism to try to argue for encryption backdoors, they're now using the threat of CSAM, @ProtonPrivacy's @andyyen tells @MORGANMEAKER https://t.co/5VqAnR34X5

2023-01-03 17:28:22 @Solivagant @Vincenzo_Natali @lisajoynolan @GreatDismal @ThePeripheralPV It wasn't sadly

2022-12-31 22:13:18 @gideonw They needed redundancy. In case one gets wiped out. Especially since we’re both bikers

2022-12-24 15:17:16 RT @marynmck: New at @WIRED, from @glichfield and me: Mpox is now just a trickle of cases in the US, and that looks like a public health su…

2022-12-21 15:54:00 @miblogestublog Hahaha…. Sorry, jajaja

2022-12-20 22:58:26 Hey media, can we please have a rule to never publish a story about a company being fined without including its annual revenue, profit or other measure of how much pain the fine will cause https://t.co/r9bP0kTibw

2022-12-20 17:31:05 TIL that Twitter is no longer blocking Mastodon links, thank you Will https://t.co/aMBMSTRhB7

2022-12-20 16:55:33 The story of what may be the first ever lawsuit based on the harm of fossil fuels to future generations. That's some #KimStanleyRobinson #MinistryForTheFuture shit right there. https://t.co/VRxdpSHsFv

2022-12-19 17:34:28 I recommend spending several minutes staring at the gif and then trying to read the story—"Physicists Rewrite a Quantum Rule That Clashes With Our Universe" https://t.co/FeQM4xiKZ1 https://t.co/S3UT20In8I

2022-12-17 17:34:01 @Ananyo That’s kinda what I’m saying. Some on the left assume Twitter will become all Nazis. But once there are no liberals left on Twitter to attack, the Nazis will get bored. I’m suggesting the core will still be right-leaning but not so extreme

2022-12-17 15:45:44 @Digidave @NiemanLab An actually interesting use case. How good would you say the summaries are?

2022-12-17 15:42:26 (Correction: can’t run for president, not born here)

2022-12-17 15:42:02 @HugoPrevost Good point, thank you

2022-12-17 15:39:13 It becomes the place you have to be to influence conservative conversation. (Again, watch out, Fox News!) Makes Musk the chief power player in conservative politics. (Watch out, Peter Thiel!) Maybe runs for president, maybe content to just run the president.

2022-12-17 15:35:22 Advertisers drift back. It becomes an OK business, not great, not justifying the $44 bn purchase price, but that doesn’t matter to Musk. He’s still ultra rich, has other companies, and has a platform he controls. Like a robber baron owning newspapers.

2022-12-17 15:32:02 My current #boringprediction for Twitter: as more liberals leave, the conflicts subside and it settles down into a fairly broad right-leaning platform - not Parler or Gab, more like the social media version of Fox News. (Watch out, Fox News!) https://t.co/PHvPf5dR70

2022-12-17 14:41:16 @cwarzel Try turning it off and then on again

2022-12-16 14:17:47 Meanwhile in the forgotten corners of the world that are not Twitter https://t.co/XrdEUVwaqd

2022-12-16 14:12:39 https://t.co/yi7IJE9pcF

2022-12-16 14:08:04 I’m fascinated by the sub-genre of tweets that find ways to rationalize Elon’s behavior as perfectly normal, nothing to see here https://t.co/NEPVYQI9gR

2022-12-16 13:47:58 Hot take: Elon’s only reason for doing ever more outrageous things is to keep Twitter numbers high. This can’t go on indefinitely though. https://t.co/IBKmIl4XWh

2022-12-16 13:39:18 @NateSilver538 I mean he didn’t do almost any of the things you’d expect a president to do when being president so

2022-12-16 13:37:25 @levie The new mission of Twitter is to be about Twitter

2022-12-16 05:17:21 @katienotopoulos It was @doomscroll_bot telling us we all need to get some sleep

2022-12-16 04:54:02 Thread of all the journalists that free-speech-loving Elon Musk who said he hoped his worst critics remain on Twitter has banned from Twitter today https://t.co/BI5PXef94O

2022-12-16 04:50:24 TFW you find all your friends having the same bad late-night screen time habits as you https://t.co/3PBJ2BL7DY

2022-12-14 21:28:48 In which, inspired by Richard Powers' "Orfeo", I offer some year-end reflections on the state of tech couched in musical terms. https://t.co/JHKLL1qpPs

2022-12-13 20:32:52 And with that said, if you ARE leaving Twitter: https://t.co/JW2KlND6wK

2022-12-13 20:30:51 @ceonyc We got you https://t.co/JW2KlND6wK

2022-12-13 20:29:08 In which @GregoryJBarber pours cold water on hot fusion https://t.co/Ab38anexcR

2022-12-13 18:27:29 Today we launched @wiredenespanol, the first-ever Spanish-language edition of @WIRED, with @urbano_hidalgo leading a small but mighty team of editors to cover the world's biggest challenges and the role of technology in tackling them https://t.co/DnLVBGQci3

2022-12-13 18:23:49 Hoy lanzamos @wiredenespanol, la primera edición hispanófona de @WIRED, bajo la dirección de @urbano_hidalgo y con un pequeño pero formidable equipo de editores, para cubrir los retos más urgentes del mundo y el papel de la tecnología en solucionarlos https://t.co/DnLVBGQci3

2022-12-13 17:22:43 I'm not leaving Twitter for now. I respect everyone for whom Musk's antics (attacks on Fauci and Yoel Roth) have crossed a line, but if we all abandon this platform it becomes what Parler or Gab only dreamed of being, and I don't want to see that in the world.

2022-12-13 13:11:50 @qntm I needed to hear it, because I hadn’t seen the collection and I loved Antimemetics Division, so thank you

2022-12-13 13:05:19 I should have added that theories about Musk having some sort of genius hidden agenda that we’re all failing to notice fall into the same category of getting misled by one’s obsession with a success-or-failure narrative https://t.co/I7eHLhZA7f

2022-12-13 12:52:07 I’m having strong deja vu about both Musk’s ever more extremist shitposting and everyone’s reactions to it https://t.co/0qeytMVoNW

2022-12-09 20:33:18 EXACTLY @amitkatwala https://t.co/TixGcnA7Yl

2022-12-08 13:00:00 CAFIAC FIX

2022-12-07 08:00:00 CAFIAC FIX

2022-11-15 01:13:34 It will turn out that the time intervals between each tweet in minutes converted to hex and combined into one string correspond to the private key for a wallet with all the money stolen from FTX https://t.co/WcM26JiZS5

2022-11-14 16:31:28 Q: "What did you make of Elon Musk poking fun of Mastodon in a recent tweet?"A: "Honestly, it was really a good thing for us. It's free advertising, and he's just making a fool of himself."Mastodon's founder on two crazy weeks https://t.co/4JgwZNUyuA

2022-11-14 05:48:34 @karenkho I feel so seen

2022-11-14 05:41:32 Shhh, he’s just playing Wordle https://t.co/aTJhNP3izx https://t.co/fqv2oLxKFN

2022-11-14 05:21:49 RT @a_greenberg: An entire industry of cryptocurrency-tracing firms now regularly tracks the loot from crypto heists and IDs culprits. Now…

2022-11-13 17:47:36 aaaaand scene https://t.co/OYVRHJyZY3

2022-11-13 05:12:09 @olivia_p_walker @WIRED I like numbers too! I can't personally vouch for this report's methodology, but in its sample of sites across a range of industries, Twitter is 10% of social traffic and social is 6.4% of all traffic. So Twitter is less than 1% of all traffic. https://t.co/yW5nkREErZ

2022-11-13 05:02:31 And @ParselySupport @parsely back me up here. Your network referrer dashboard seems to be down (tried multiple browsers and devices) https://t.co/Dg0WTmMZCO

2022-11-13 05:00:39 @olivia_p_walker @WIRED How about we have a convo about the literal meaning of the words “biggest click driver on the internet by far”

2022-11-13 04:35:29 Hahahahahahahaha no. Less than 2% of @WIRED traffic https://t.co/bdVmdIIoJP

2022-11-12 21:04:31 Listening to all these people debating whether a police car outside a building means SBF has been arrested and tbh I’m glad they’re learning that citizen journalism is hard https://t.co/6RC9H0SL1k

2022-11-11 23:14:02 Ok now I know that all I need to do to get my team working harder is to repeat the words “maniacal sense of urgency” at them several times in quick succession. Like casting a spell. https://t.co/E7TKjn15cC

2022-11-11 17:02:57 @alexhern https://t.co/M9zHYmU8sw

2022-11-11 16:48:07 Get ready for the Great Reinvention https://t.co/0YAGxDv7od

2022-11-11 04:23:17 So is Official now back Officially? I can’t keep up. https://t.co/HVfvMiIeVq

2022-11-10 21:14:20 Meta's massive layoffs weren't just the result of bulking up too quickly to build the metaverse, but 5-10 years of failed bets, a recent employee tells @stokel https://t.co/XnjQcDfZeD

2022-11-09 18:38:00 Musk on the new verification criteria for the blue checkmark: "Someone has to have a phone, a credit card, and $8 a month. That's bar. However, we will actively suspend accounts that engage in deception or trickery." https://t.co/6ZA6Z8Rgxf

2022-11-08 02:01:53 By this point I'll wager they have also discussed charging people to tweet but not to read, putting blue checkmarks on the blockchain, selling it to Meta, and selling it to the Russians https://t.co/VzgrmahHPu

2022-11-08 01:47:07 This is like all those people who used to say Trump just needed some adults in the room https://t.co/FyZytQlCLY

2022-11-01 17:48:27 I'm on the fence about whether I'd pay for Twitter verification (and at what price) but posting this as insurance https://t.co/96wfCBZ9qa

2022-11-01 15:01:31 @ftrain And they're released in stages - 5,000 the first year, 2,500 the second, 1,250 the third...

2022-10-27 21:55:54 @AndrewCouts gruey

2022-10-27 20:16:42 Someone please remind me what web4 was, I must have blinked and missed it https://t.co/navLqsD8r1

2022-10-27 20:12:28 https://t.co/qvtD6XEziW

2022-10-27 20:12:09 @bellye66 https://t.co/qvtD6XnwgW

2022-10-27 20:11:45 @DigitalJonathan more a case of where ARE the conversations https://t.co/qvtD6XmYro

2022-10-27 20:10:24 @fxshaw Discord, Slack, Reddit, Telegram, WhatsApp &

2022-10-27 19:59:09 RT @emilylmullin: Happy publication day to my colleague @mrMattSimon, whose book on how microplastics have infiltrated our bodies and the p…

2022-10-27 19:56:33 My Twitter prediction, fwiw: not much change. Trump et al will come back, trolling will increase somewhat, rest of us will block and mute more and engage less but still use it for publishing—more web 1.0, less 2.0. Real conversation is already happening elsewhere.

2022-10-27 19:51:18 Tech industry earnings got you down? Worry not: @WIRED is hiring! Check out our three newly posted jobs: service editor, deputy ideas editor, and executive editor, newshttps://t.co/WLEuNoqdhehttps://t.co/RYrdKV7APvhttps://t.co/bKJA468F4kPlus more: https://t.co/MTz4iKAZaG

2022-10-25 20:44:18 @AndrewCouts @WIRED https://t.co/kbLXwVMsM1

2022-10-25 18:45:41 @CaseyNewton 21st century global queer subcultures, ranked in order of preciousness

2022-10-25 18:38:18 @WIRED @a_greenberg (If you can't quite face reading 40,000 words on crypto by @matt_levine, I can thoroughly recommend 30,000 words on crypto crime by @a_greenberg)

2022-10-25 18:37:00 Today @WIRED launches the first in a six-part series: the epic story of the takedown of AlphaBay, the dark web's largest ever criminal marketplace, by @a_greenberg (from his upcoming book, TRACERS IN THE DARK). Hi-tech detective non-fiction at its finest. https://t.co/uk4XzqXQx9

2022-10-20 14:29:03 @karenkho Are you roomscrolling

2022-10-20 13:08:04 At this rate King Charles is going to get through more prime ministers in his life than his mother did

2022-10-20 13:06:33 Trusst fall

2022-10-18 21:07:24 The clocks on the website are the id of @semafor. DC, Brussels, Lagos, Beijing, Singapore

2022-10-17 16:49:43 "Few things can turn around a boring winter afternoon like receiving a billion dollars. The amount is there, in its 10-figure glory, glaring from my phone screen in white characters against a charcoal background: $1,112,172,834."@Gmvolpi's last @WIREDhttps://t.co/S7LhY3beoq

2022-10-16 21:35:28 All photos by @joepugliese for @WIRED, dogs (mostly) via @Muttville #SeniorDogRescue

2022-10-16 21:31:48 Uncle Grady, 13, HavaneseLifespan: 14–16He has a strong preference for people food. https://t.co/45dGrjA5Gb

2022-10-16 21:31:02 Bumper, 8, American Staffordshire terrierLifespan: 13–15Recently adopted, Bumper has a tail that wags a mile a minute. https://t.co/jlqIFHSxPd

2022-10-16 21:30:12 Trout, age 4, a 21-breed “supermutt” Lifespan: 12+Hailing from the streets of San Francisco's Mission District, he apparently walked himself to a shelter. https://t.co/Cy5ja63Hf0

2022-10-16 21:29:05 Corky, 11, PugLifespan: 13–15If she thinks you're not paying enough attention to her, Corky will tap you with her paw and sneeze on you. https://t.co/xb0klxEBu4

2022-10-16 21:28:11 Peanut, 7, Mini-poodle/Chihuahua/Pomeranian mixLifespan: UnclearWhenever her dog friends visit her home, she brings out all her toys and piles them up for her buddies to see. https://t.co/uEfeNCM0IM

2022-10-16 21:27:08 Luna, 10, French bulldogLifespan: 10–12Despite being blind, she always finds a perch on the highest available pillow. https://t.co/jrxexvqUf6

2022-10-16 21:26:26 Odin, 4, Great DaneLifespan: 7–10His favorite toy is Harry Elephante, a stuffed elephant with crinkly ears. https://t.co/Ovrw9HKi7e

2022-10-16 21:25:28 Smitty &

2022-10-16 21:22:59 Name: Penny. Age: 14. Breed: Greyhound. Lifespan: 10–13 years. Trained as a professional racer, she competed in five races and never left the gate, so she was fired and picked up by a rescue organization. https://t.co/KzEqZXvHz6

2022-10-16 21:22:02 Here is a story about how life-extension treatments for dogs are on the way, and how this might lead to similar treatments for people. Also, it has lots of delightful portraits of dogs.https://t.co/yHpm2rDNqV

2022-10-16 20:33:51 In which I test-ride the latest electric Harley and am left wanting more https://t.co/KMPRqUd2Uo

2022-10-15 16:53:15 Achievement unlocked https://t.co/uDEFPzQ163

2022-10-14 19:31:23 A sci-fi alternative history in which the entire US government had been legally forbidden from using computers since the 1980s

2022-10-13 21:59:58 One study, two stories https://t.co/U4lXRPRgEv

2022-10-06 19:56:01 Gonna say it again: I think this kind of AI-generated video is going to open up an entire new field of possibility in journalism, where you report stories and then "write" them in video form https://t.co/cIwwTkZfGe

2022-10-04 20:05:29 Yeah, so how come @cwarzel uses AI-generated art once and has to make a groveling apology to the internet, but @CaseyNewton uses it for months and nobody complains? Not a criticism of you, Casey, rather of Twitter's double standards https://t.co/UmIxSPemhc

2022-10-04 18:30:16 Congrats to @GregoryJBarber for winning @theAGU's Walter Sullivan award https://t.co/Nxb5GtC91k - read his winning story here https://t.co/hwg3SMyCi0

2022-10-04 13:29:13 Welcome to @WIRED, @JKFruit! https://t.co/DOemaxuCzA

2022-10-03 13:27:17 "In all, finding Twitter’s secretive bot squad takes me two months and dozens of interviews with people living in six countries across three different continents..." @morganmeaker goes on an odyssey https://t.co/dbvru2G8tk

2022-10-03 12:54:57 In which @LaurenGoode tries to separate the bots from the boys https://t.co/LsSuLWmfh3

2022-10-02 16:52:50 @liron Just to be clear, this is not like that time you started making pro-web3 arguments as an ideological Turing test, right? You actually believe this?

2022-09-30 00:57:23 RT @LaurenGoode: What comes after The Merge? How should crypto be regulated? And what would have to happen for @VitalikButerin to conclude…

2022-09-29 15:59:15 Of course, hugely problematic because creating artificial video creates details of a scene that were never in the information you reported. How is the viewer to tell what’s actual info and what’s interpolation? But I suspect conventions will emerge.

2022-09-29 15:56:14 @stokel Welcome to the rest of your life

2022-09-29 15:54:33 Yup. It was only a matter of time before this happened. I’ve been saying for a while that this could revolutionize journalism. What if instead of writing a story you’ve reported, you “write” video of it instead? What new audiences might you reach? https://t.co/YNSfeHCgnT

2022-09-29 15:41:33 @EvanSelinger That “democratizing” a technology is a way to reduce power imbalances or inequities (as opposed to just recreating them in another setting)

2022-09-28 17:27:39 At #REWIREDGreen @kenlacovara kicks it off with a riveting reminder of what’s at stake: a species probably went extinct while he was speaking https://t.co/JQe7y82uZI

2022-09-24 17:48:28 Lisa has been trying to get a response from Twitter with no luck, so if @TwitterSupport is reading this, please reach out to me. And if you see offers of NFTs or anything else on @lisang, don’t click on them - you’ll be scammed. See previous similar cases https://t.co/UGfgt4U0Bk

2022-09-24 17:45:59 According to Lisa, who I’m in touch with, “I got a message at 9.30 am that a user in Moscow had tried to access my account. I logged in right away but it was too late … Also got a message that the hacker had switched off two-factor authentication, which I find incredible”

2022-09-24 17:44:13 This “Ethereum” account is actually a verified account belonging to the journalist Lisa Goldman (@lisang). This morning it was hacked and taken over. It’s been retweeting a stream of crypto content to hide its origins. You can see in the second screenshot when it switched over https://t.co/pJPQFqUOg9

2022-09-22 15:29:55 Over the past year or so I've noticed that most PR emails I get no longer have an unsubscribe link. Can anyone in the PR industry tell me why this is? (FWIW, the only effect this is having is that I'm now blacklisting email addresses instead of clicking unsubscribe.)

2022-09-22 15:17:19 Ayer el #WIREDSummit en la Ciudad de México fue todo un éxito, con tres escenarios, decenas de ponentes y unos 1,500 visitantes -- uno de los eventos más ambiciosos que @WIRED jamás ha montado en cualquier país. Ojalá sea solo el primero de muchos! https://t.co/PuYTkQj2Vi

2022-09-22 15:10:58 RT @PublimetroMX: #PublimetroMX tuvo una entrevista exclusiva con Gideon Lichfield, director Editorial Global de Wired, para hablar de la i…

2022-09-15 15:39:33 RT @AndyFidel_: Are you based in Mexico City (#CDMX)? Join @CathyHackl at #WIREDSummit next week where she will be doing a keynote followed…

2022-09-12 21:28:12 One of my earliest stories as a science reporter 25 years ago was about exoplanets, which were just starting to show up as minute shifts in stars' emission spectra or brightness. That we can now actually take a picture of one is mind-boggling. https://t.co/huG2wG9q4L

2022-09-12 21:23:31 I used to think I might be immune to Covid. I'd been going out and traveling and attending conferences and never gotten sick... until I did. But there's now a big enough sample of people with possible natural immunity to do some science. By @gracefbrowne https://t.co/GEQnarFr9I

2022-09-09 02:04:12 FFS ⁦@BBC⁩, can’t you get punctuation right just this one time https://t.co/yfx7viOkEl

2022-09-08 18:45:05 @natashaloder @stokel @munkeatlooi Oh, he's fast. I think of Chris as running a kind of overclocked model of reality in his head so he can figure out what the story is going to be about an hour before it happens

2022-09-08 18:28:13 On the queen as internet phenomenon. "You can kind of read into her whatever you want to read..." "The queen is a vehicle for adding memetic subjectivity..." "Memes thrive because of communal experience—something we can all share and relate to." by @stokel https://t.co/XG0kQIAaIo

2022-09-07 18:35:23 @stokel +1 on @BostonJoan

2022-09-02 02:39:51 eyerolls

2022-09-01 20:26:37 When US forces left Afghanistan one year ago, tens of thousands of Afghans who had worked for them were left in limbo and in danger. This is the extraordinary story of the volunteers who pulled together to get them out https://t.co/m7HqBjPirv

2022-08-31 00:20:38 @codybrown @matthewroche Can recommend another Richard Powers: Orfeo, which I feel also intersects philosophically and spiritually with those three you listed

2022-08-25 22:16:31 RT @thu: huge wondrous wild honor to talk to @sayakamurata about the struggles of having a uterus and being a human.my profile is on @WIR…

2022-08-23 23:20:10 @karenkho I entered my 40s, realized life was only half over and breathed a sigh of relief because there was so damn much I still hadn’t done

2022-08-22 15:33:01 "'When you see trees growing, you know that the climate has really shifted,' says Dial. 'It's not like five years of weather, or 10 years of weather. It's 30 years of climate that's established new trees in new places.'"@mrMattSimon on Arctic greening https://t.co/NujnTMbyuP

2022-08-18 16:48:15 The Merge is finally happeninghttps://t.co/WWK23y4bSU

2022-08-17 11:39:27 A history of “authenticity”, how every single social media app has tried and failed to conjure it, and why BeReal—whose entire identity rests on the concept—will be no exception, because “authenticity” was a marketing tool from the very beginning https://t.co/oNAECQrVno

2022-08-15 03:52:13 @lanegreene @zseward And it took @TheEconomist only 10 years to catch up. What heady times we live in.

2022-08-10 21:03:01 @evacide And what are you, the person assembling the team?

2022-08-10 20:57:34 How will monkeypox play out? Having declared it a public-health emergency, the US (and the world) is now at an inflection point. @marynmck looks at the scenarios https://t.co/gFQAmpcvg7

2022-08-08 21:23:58 @zakjason What kind of behavior do you model in the Jason household

2022-08-08 19:29:53 "To understand the fight over regulating crypto, it helps to start with the orange business." @GiladEdelman on the coming shootout between the US government and the crypto industry https://t.co/CrOVNuVCVz

2022-08-08 15:37:18 Why are there 16m doses of monkeypox vaccine in the world, more than enough to contain the outbreak, and yet people can't get shots? "What Covid—and now monkeypox—reveals is that tackling outbreaks as a domestic problem simply does not work." @gracefbrowne https://t.co/SAL4gn8ude

2022-08-08 15:22:29 I mean, if you're really going to follow the PR strategy of spamming journalists with automated emails at least learn to use your spam software correctly https://t.co/uWs3bNNHYp

2022-08-04 14:19:18 Where did covid come from? A scientific detective story, by @amitkatwala https://t.co/o2wM0YMdgj

2022-08-04 14:07:53 actually forgot the other koan https://t.co/CONKZSSkuC

2022-08-04 14:06:42 My Twitter feed today is koans https://t.co/WJO2hJcAYy

2022-08-03 21:20:53 @CyrilPedia @adamdavidson Oh man. Where to begin. The launch costs. The targeting systems. The kashrut certification.

2022-08-03 21:20:02 @gmochkofsky You haven't aged a bit

2022-08-03 21:16:11 You see, when you stop treating it as a conspiracy theory and instead discuss it as a logistical problem, it becomes waaaay easier to dismiss https://t.co/FbTpcOtQMB

2022-08-03 21:13:52 Just got out of a planning meeting for RE:WIRED Green, our event on sustainability &

2022-08-03 11:10:20 RT @JayDProsser: 'It was possibly the worst drag performance I had ever seen, but it didn’t matter. It was also an act of sheer, joyous def…

2022-08-03 11:09:25 I met @gmochkofsky 17 years ago. We visited the West Bank together to meet the unlikeliest of all Israeli settlers, a community of converts from the mountains of Peru. Now her book on their strange origin story is finally out in English https://t.co/aWqsUSJJ6n

2022-08-01 15:49:45 RT @snackfight: WIRED is hiring! Come work with me and the dynamic, award-winning team of journalists here. We're looking to fill multiple…

2022-07-24 16:25:50 @liron As @moxie said, people don’t want to run servers. They’ll rely on centralized exchanges, hosting services for dApps and NFTs, etc to talk to any blockchain. So the supposed benefits of decentralization will never actually apply for most people. https://t.co/Rzn3T1LIaQ

2022-07-22 03:29:26 @thegarance Yep, I was one of those holdouts until a week ago

2022-07-22 02:39:16 The fires sweeping Europe aren’t just a result of climate change but a complex set of factors including migration, fire management policies, and demographics, as @mrMattSimon explains https://t.co/XNM8sE9X7f

2022-07-21 21:29:52 Didn't expect to get sucked in to a move-by-move breakdown of a NASCAR pit stop but this is nerd heaven. Every tenth of a second is choreographed. Everything from the position of a crew member's hands on a wheel to the angle of their hips is pre-planned. https://t.co/EaDB58QrL0

2022-07-21 21:17:26 @LaurenGoode My hypothesis is that PR folks got beta access to GPT-3 and have just been using it ever since to write press releases without telling their bosses, since they know the response rate is going to be next to zero anyway

2022-07-21 12:32:51 @niubi idk, if I were paying a service exposing people who are full of shit only to find it caused them to gain even more support, I’d probably ask for my money back

2022-07-20 20:17:27 @Knibbs you have to have the salt flakes or it doesn't work

2022-07-20 16:50:03 The story behind the MoonSwatch and its "Bioceramic" body, which is neither biodegradable nor a ceramic https://t.co/jboB19cBud

2022-07-20 14:54:05 @elipariser I think this is like "hearts and minds" or "thoughts and prayers." Basically, things sound more meaningful in pairs, so you find a pair of things for which it would seem strange to want one without the other

2022-07-20 14:08:15 “You’re mad! That’s the maddest thing I’ve seen all day!”If you're a #Brompton fan (and also if you're not) you'll love @miss_eleneous's hilarious account of her attempt to ride the titanium-framed, £4k/$5k T-line through 70 miles of mud and gravel. https://t.co/p2A4jQpowE

2022-07-19 22:32:19 I was wondering this just a couple of days ago, and boom, here it is as @emilylmullin's first story as a @WIRED staff writer. Editing by telepathy has been achieved https://t.co/DegEUkGley

2022-07-18 23:45:19 @evelyndouek Congratulations!

2022-07-13 13:08:46 @ebottcher @CMChiOsse @DMCarr The problem isn't that the portal crashed. It's that the portal is the only way to get appts, so only people like me with desk jobs can grab one. Use your hearing to make @NYCHealthCommr explain why his dept's approach to public health is so inequitable. https://t.co/mu71vpbasV

2022-07-13 13:08:14 @NYCHealthCommr How many doses have you made available via community-based organizations, vs through the @nycHealthy online portal?

2022-07-12 20:57:02 Building a stable appointment infrastructure will merely make it quicker for the very few people who can get an appointment to get one. It won't solve the problem of who these appointments are going to and how to have the best public-health outcome https://t.co/G0bwegvAB2

2022-07-12 20:54:00 No, it is not just further proof that demand is very high. It is further proof that this is not the right way to distribute something when demand outstrips supply by one or two orders of magnitude. https://t.co/8K157fcgt6

2022-07-12 17:45:53 RT @TanmoyDasLala: @nycHealthy how is this an equitable approach? what about our folks who are in jobs, with clients, who can't refresh all…

2022-07-12 17:31:37 The problem is not that the site can't handle the traffic, the problem is this should never have been the method of distributing appointments in the first place https://t.co/AkiR2hiDsa

2022-07-12 17:30:19 If @nycHealthy had even a shred of common sense it would have figured out the sensible and equitable way to distribute limited vaccine supplies to those most in need and whose vaccination would be most likely to slow the outbreak, rather than those with the most privilege

2022-07-12 17:26:43 This. It was entirely predictable that in a city with 100,000s of gay men, releasing 1,250 monkeypox vaccine appointments at a pre-announced time would crash the servers. It's also grossly inequitable, favoring ppl like me who can sit at a desk during the workday hitting refresh https://t.co/v37RdA8AJn

2022-07-08 19:41:33 RT @brbarrett: Cruise runs a fleet of robotaxis in SF and has lost comms with them multiple times, resulting in chaooosssssssss as they get…

2022-07-07 10:46:49 @jetjocko That one’s for when she fired Prince Andrew

2022-07-07 10:41:21 The Economist goes to press late Thursday morning, around exactly the time Johnson quit. So yes, this timing is incredible. Was that the cover already or did they scramble to do a new one, @tom_nuttall ? https://t.co/WDTjXXF3is

2022-07-07 10:27:38 RT @KarlreMarks: It's great that the British are discovering how difficult it is to get rid of British rule

2022-07-07 02:09:41 @jetjocko Actually it’s the theme music from Chariots of Fire but that’s OK, we can do Highlander on the dubbed-for-Americans version

2022-07-06 23:00:03 A clusterfuck in six tweets https://t.co/0YxU0j44Hn

2022-07-06 14:30:15 RT @GregWilliams718: Yup, definitely ignore the 'clever people' responsible for a global programme that successfully eradicated smallpox. D…

2022-07-06 14:11:57 It's a moral dilemma. "We can’t go around crying all the time. And yet I hope these parents and their children still find people to bear witness. There must be a solution beyond pretending tragedy can be disappeared by fiddling with algorithms." https://t.co/MEuqTnYr5Z

2022-07-06 14:09:20 "If I could hit one button to hide all content relating to ill or dead children, I would," writes @Knibbs. "But there’s also something uncomfortable about the idea that the way to fix this is to hide things that grieving, scared parents are trying to share about their families."

2022-07-06 14:08:02 It's a familiar story by now—the algorithmic cruelty of social media, flooding people with content that traumatizes or re-traumatizes them because it matches things they've viewed before. (See @LaurenGoode's unforgettable story about her wedding https://t.co/htLIPhA2rd

2022-07-06 14:05:49 Since giving birth to her son, @Knibbs's social media feeds have filled up with posts "about babies and children who are ill, dying, or dead." https://t.co/MEuqTnYr5Z

2022-07-06 00:23:56 @karenkho Somehow I never see your viral tweets but always your tweets about going viral

2022-07-05 22:33:34 All this https://t.co/YhCwRT9Yq5

2022-07-05 16:42:11 RT @evacide: The difference between now and the last time that abortion was illegal in the United States is that we live in an era of unpre…

2022-07-05 16:40:50 Welcome to the team, @emilylmullin! Looking forward to your explorations of the wild frontiers of biotech for @WIRED. https://t.co/Iybqb5JVNn

2022-07-05 16:39:01 If you've been struggling (like me) to make sense of, uh, ~what the hell is happening~ in US and EU crypto regulation, here's @Gmvolpi with a great overview of the securities v. commodities debate, the likely fate of stablecoins and exchanges, and more https://t.co/xSGrRdiwIk

2022-06-30 15:36:39 This seems to be a logic common to both the Dobbs ruling and today's EPA ruling: to force Congress to stop delegating (to courts in the first instance, and agencies in the second), and "do its job". Hard to see how this does anything but totally break the US government. https://t.co/5fNql1i4Zp

2022-06-28 15:30:15 Just nine years after moving to the US from Argentina, @gmochkofsky has already founded a journalism program, become a @NewYorker columnist, published a book in English, and now been appointed dean of @newmarkjschool at @CUNY. Do not underestimate her. ¡Felicidades Graciela! https://t.co/N0NYtuSdLh

2022-06-27 20:55:55 As the world moves towards EVs, China is poised to dominate the industry the way it has dominated electronics manufacturing. From @willknight, the first in a series https://t.co/g324OeG9q8

2022-06-27 20:53:06 @jtemple but you had the presence of mind to take an awesome selfie which is how we know you're doing just fine

2022-06-24 14:53:48 With Roe v Wade overturned, it's time for people in states that will ban abortion to start thinking about what to do in the event that they ever need one https://t.co/NqkBjd6dsd

2022-06-24 14:51:33 One of the overlooked issues in forcing unwanted pregnancies is that pregnancy itself can be medically risky, and banning abortion will cause more mothers to die, including some time ~after~ their children have been born https://t.co/VpFphDbz7L

2022-06-24 14:41:54 A look back at what Roe v Wade changed for women, and what will change now that it's been overturned, by @marynmck https://t.co/6foqF23DlQ https://t.co/LBpjoP3r4Q

2022-06-24 14:30:38 To be clear, Clarence Thomas is suggesting the Supreme Court should relitigate the right to contraception, private consensual sexual acts, and same-sex marriage. https://t.co/vg71SyOlor

2022-06-23 20:05:44 @justinhendrix @JeffHorwitz @mims @BritishTim And to @jonathanstray

2022-06-23 17:49:22 "This is roughly the point in the worst-case scenario when the meltdowns at nuclear power plants begin."Read @MatthewSRibel's sobering account of what it will be like to live through The Next Big One... from the Sun. https://t.co/hxmtJEwOcV

2022-06-23 17:47:47 "Heavy medical hardware—dialysis machines, imaging devices, ventilators—will cease to function, and hospital wards will resemble field clinics. With death tolls mounting and morgues losing refrigeration, municipalities will face grave decisions about how to safely handle bodies."

2022-06-23 17:47:31 "As the outage goes on, health care facilities will grow overwhelmed. Sterile supplies will run low, and caseloads will soar. When backup batteries and generators fail or run out of power, perishable medications like insulin will spoil."

2022-06-23 17:46:57 "Homes and offices will lose heating and cooling

2022-06-23 17:46:17 The largest transformers...are firmly anchored into the ground, using Earth’s crust as a sink for excess voltage. But during a geomagnetic storm, that sink becomes a source... Storm-induced direct current can cause them to overheat, melt, and even ignite." https://t.co/hxmtJEwOcV

2022-06-23 17:44:12 "Through a weird... property of electromagnetism, the electricity coursing through the atmosphere will begin to induce currents at Earth’s surface. As those currents race through the crust, they will seek the path of least resistance... upward, through the electrical grid."

2022-06-23 17:42:58 "When navigation... systems fail, the commercial airline fleet—about 10,000 planes in the sky at any given time—will attempt a simultaneous grounding. Pilots will eyeball themselves into a flight pattern while air traffic controllers use light signals to guide the planes in."

2022-06-23 17:41:29 "As the atmosphere heats up, it will swell, and satellites will drag, veer off course, and risk collision with each other and space debris. Some will fall out of orbit entirely." https://t.co/hxmtJEwOcV

2022-06-23 17:40:55 "The plasma will... flood Earth’s ionosphere, and the electron bombardment will cause high-frequency radio to go dark. GPS signals, which are transmitted via radio waves, will fade with it. Cell phone reception zones will shrink

2022-06-23 17:40:07 "When the next solar storm approaches Earth... alarms will sound on crewed spacecraft. Astronauts will proceed to cramped modules lined with hydrogen-rich materials like polyethylene... They may float inside for hours or days." https://t.co/hxmtJEwOcV

2022-06-23 15:48:35 ...but that led to a glut of supply, now made worse by inflation and supply chain bottlenecks, and since then freight rates have come down by a third. "I've never seen rates falling in June."

2022-06-23 15:46:46 Fascinating insight into cyclicality of trucking from @daniellewis at #Collision2022: N Am truck freight rates roughly doubled in the 18 months to last December. Many truckers left large firms to start their own: up to 10,000 new trucking companies being registered every week... https://t.co/CVDYPBaiCa

2022-06-21 19:06:13 @MalwareTechBlog @klrgrz Did the same to me the first time around. Seriously unpleasant. Was quite helpful the second time. But I had a specific anxiety the first time that I didn’t the second, so I concluded it just exacerbated that. Also, you might habituate to it after a couple of days.

2022-06-20 22:23:03 In 40 minutes I’ll be interviewing @MargaretAtwood and @YungWu on stage at @CollisionHQ #Collision2022 about climate change, the influence of art on tech, and reasons for optimism - live stream at https://t.co/9ei0w4RfvQ

2022-06-20 14:05:35 RT @CollisionHQ: Who’s ready to kick off #CollisionConf 2022!? We’ll be joined by a lineup of incredible speakers for Opening Night incl…

2022-06-18 09:43:09 You should read this ⁦⁦@StevenLevy⁩ interview with the Google engineer who believes an AI is sentient, because (a) it’s fascinating and (b) it’s a model of how to interview someone critically yet non-judgmentally, which few people know how to do https://t.co/3C7oDcpIkF

2022-06-17 11:41:45 @BDUTT @lokkalyandas @pinarayivijayan Congratulations Barkha, and "Czarina of the Digital Days" is an excellent title

2022-06-17 07:51:10 I’m on the @mediamasters_fm podcast talking about the future of @WIRED and the media in general https://t.co/7bT2eAGYcH

2022-06-15 22:08:30 RT @mrMattSimon: Microplastics have infested every corner of Earth, from the highest mountains to the deepest seas. They’re in our food, wa…

2022-06-15 22:08:17 @jonathanstray Aren’t I objectively not allowed to have a favorite?

2022-06-15 08:53:39 @MedinaMora Yo que sí, tan solo porque @zeynep es antropóloga y eso le ha dado una perspectiva única en otros campos desde la epidemiología hasta el análisis del poder tecnológico

2022-06-14 16:16:15 @Chanders @jbenton Josh showed you his so I'll show you mine https://t.co/7oVct5J0pL

2022-06-14 16:11:52 For Pride month, I'm proud to present @marynmck's story on how the LGBTQ community in Provincetown helped turn last summer's Covid outbreak into a model for how public health emergencies should be handled https://t.co/14UWbZbOSs

2022-06-13 13:03:37 @Chanders Please have them analyze Barlow's Declaration of Independence of Cyberspace

2022-06-13 13:01:22 A wonderful story of Soviet-era steganographic spying from @lilyhnewman https://t.co/YPpV8PRdJe

2022-06-08 19:20:42 OK, world: what is the single most important story @WIRED should report on next year?

2022-06-06 16:58:13 RT @a_greenberg: When AlphaBay, the dark web market for drugs and fraud, suddenly reappeared last August—four years after it was torn offli…

2022-06-03 20:56:58 Aaand the $1m linchpin sponsor of a system for detecting giant asteroids before they hit Earth is… @TitosVodka. I guess there wouldn’t be many drinkers left if an asteroid wipes us out https://t.co/eGRjUfLJIY (by @raminskibba)

2022-06-02 20:51:41 RT @WIRED: In this WIRED exclusive, the president of Ukraine @ZelenskyyUa discusses what Big Tech can do to open the flow of information in…

2022-06-01 18:27:41 @pardesoteric Please post the most cravenly opportunistic ones here

2022-06-01 18:25:06 @MadelineAshby @mitpress That anthology came about because in the dark days of April 2020 I thought the world needed some hopeful fiction about what society might have learned from the pandemic. Interesting to hear from @sim_kern that other places were trying to avoid the subject https://t.co/VbztYBdHon

2022-06-01 16:36:54 Congratulations to @Megan_Nicolett on the first of many... https://t.co/fLZOFsmhSQ

2022-05-27 17:20:50 Power of the gun lobby in one crazy chart (from 2017) https://t.co/k3iEYnCjIf https://t.co/dZkgezrxuV

2022-05-27 17:12:53 @LaurenGoode Please explain how you do this

2022-05-27 17:11:23 4chan is often home to screeds by American mass murderers. But who owns it? A @Justin_Ling investigation for @WIRED reveals a close relationship with a Japanese toy firm that does deals with some of the world’s biggest entertainment companies https://t.co/WWANXj4PTY

2022-05-27 00:49:15 This is horrible, but TIL that you can literally die of a broken heart: grief can release stress hormones that stop the heart from functioning https://t.co/X9xBS7l2XF

2022-05-25 22:38:56 @amywebb Good if you are looking for a polemic, by Yanis Varoufakis https://t.co/C4MDdCsi8G

2022-05-23 22:52:49 Welcome to @WIRED, @Megan_Nicolett! Can't wait to see what you produce. (And no, that isn't a coded way of saying your first deadline is tomorrow morning. You've got at least until the afternoon.) https://t.co/uzVoG1p5dk

2022-05-20 08:11:00 CAFIAC FIX

2022-10-27 21:55:54 @AndrewCouts gruey

2022-10-27 20:16:42 Someone please remind me what web4 was, I must have blinked and missed it https://t.co/navLqsD8r1

2022-10-27 20:12:28 https://t.co/qvtD6XEziW

2022-10-27 20:12:09 @bellye66 https://t.co/qvtD6XnwgW

2022-10-27 20:11:45 @DigitalJonathan more a case of where ARE the conversations https://t.co/qvtD6XmYro

2022-10-27 20:10:24 @fxshaw Discord, Slack, Reddit, Telegram, WhatsApp &

2022-10-27 19:59:09 RT @emilylmullin: Happy publication day to my colleague @mrMattSimon, whose book on how microplastics have infiltrated our bodies and the p…

2022-10-27 19:56:33 My Twitter prediction, fwiw: not much change. Trump et al will come back, trolling will increase somewhat, rest of us will block and mute more and engage less but still use it for publishing—more web 1.0, less 2.0. Real conversation is already happening elsewhere.

2022-10-27 19:51:18 Tech industry earnings got you down? Worry not: @WIRED is hiring! Check out our three newly posted jobs: service editor, deputy ideas editor, and executive editor, newshttps://t.co/WLEuNoqdhehttps://t.co/RYrdKV7APvhttps://t.co/bKJA468F4kPlus more: https://t.co/MTz4iKAZaG

2022-10-25 20:44:18 @AndrewCouts @WIRED https://t.co/kbLXwVMsM1

2022-10-25 18:45:41 @CaseyNewton 21st century global queer subcultures, ranked in order of preciousness

2022-10-25 18:38:18 @WIRED @a_greenberg (If you can't quite face reading 40,000 words on crypto by @matt_levine, I can thoroughly recommend 30,000 words on crypto crime by @a_greenberg)

2022-10-25 18:37:00 Today @WIRED launches the first in a six-part series: the epic story of the takedown of AlphaBay, the dark web's largest ever criminal marketplace, by @a_greenberg (from his upcoming book, TRACERS IN THE DARK). Hi-tech detective non-fiction at its finest. https://t.co/uk4XzqXQx9

2022-10-27 21:55:54 @AndrewCouts gruey

2022-10-27 20:16:42 Someone please remind me what web4 was, I must have blinked and missed it https://t.co/navLqsD8r1

2022-10-27 20:12:28 https://t.co/qvtD6XEziW

2022-10-27 20:12:09 @bellye66 https://t.co/qvtD6XnwgW

2022-10-27 20:11:45 @DigitalJonathan more a case of where ARE the conversations https://t.co/qvtD6XmYro

2022-10-27 20:10:24 @fxshaw Discord, Slack, Reddit, Telegram, WhatsApp &

2022-10-27 19:59:09 RT @emilylmullin: Happy publication day to my colleague @mrMattSimon, whose book on how microplastics have infiltrated our bodies and the p…

2022-10-27 19:56:33 My Twitter prediction, fwiw: not much change. Trump et al will come back, trolling will increase somewhat, rest of us will block and mute more and engage less but still use it for publishing—more web 1.0, less 2.0. Real conversation is already happening elsewhere.

2022-10-27 19:51:18 Tech industry earnings got you down? Worry not: @WIRED is hiring! Check out our three newly posted jobs: service editor, deputy ideas editor, and executive editor, newshttps://t.co/WLEuNoqdhehttps://t.co/RYrdKV7APvhttps://t.co/bKJA468F4kPlus more: https://t.co/MTz4iKAZaG

2022-10-25 20:44:18 @AndrewCouts @WIRED https://t.co/kbLXwVMsM1

2022-10-25 18:45:41 @CaseyNewton 21st century global queer subcultures, ranked in order of preciousness

2022-10-25 18:38:18 @WIRED @a_greenberg (If you can't quite face reading 40,000 words on crypto by @matt_levine, I can thoroughly recommend 30,000 words on crypto crime by @a_greenberg)

2022-10-25 18:37:00 Today @WIRED launches the first in a six-part series: the epic story of the takedown of AlphaBay, the dark web's largest ever criminal marketplace, by @a_greenberg (from his upcoming book, TRACERS IN THE DARK). Hi-tech detective non-fiction at its finest. https://t.co/uk4XzqXQx9

2022-10-27 21:55:54 @AndrewCouts gruey

2022-10-27 20:16:42 Someone please remind me what web4 was, I must have blinked and missed it https://t.co/navLqsD8r1

2022-10-27 20:12:28 https://t.co/qvtD6XEziW

2022-10-27 20:12:09 @bellye66 https://t.co/qvtD6XnwgW

2022-10-27 20:11:45 @DigitalJonathan more a case of where ARE the conversations https://t.co/qvtD6XmYro

2022-10-27 20:10:24 @fxshaw Discord, Slack, Reddit, Telegram, WhatsApp &

2022-10-27 19:59:09 RT @emilylmullin: Happy publication day to my colleague @mrMattSimon, whose book on how microplastics have infiltrated our bodies and the p…

2022-10-27 19:56:33 My Twitter prediction, fwiw: not much change. Trump et al will come back, trolling will increase somewhat, rest of us will block and mute more and engage less but still use it for publishing—more web 1.0, less 2.0. Real conversation is already happening elsewhere.

2022-10-27 19:51:18 Tech industry earnings got you down? Worry not: @WIRED is hiring! Check out our three newly posted jobs: service editor, deputy ideas editor, and executive editor, newshttps://t.co/WLEuNoqdhehttps://t.co/RYrdKV7APvhttps://t.co/bKJA468F4kPlus more: https://t.co/MTz4iKAZaG

2022-10-25 20:44:18 @AndrewCouts @WIRED https://t.co/kbLXwVMsM1

2022-10-25 18:45:41 @CaseyNewton 21st century global queer subcultures, ranked in order of preciousness

2022-10-25 18:38:18 @WIRED @a_greenberg (If you can't quite face reading 40,000 words on crypto by @matt_levine, I can thoroughly recommend 30,000 words on crypto crime by @a_greenberg)

2022-10-25 18:37:00 Today @WIRED launches the first in a six-part series: the epic story of the takedown of AlphaBay, the dark web's largest ever criminal marketplace, by @a_greenberg (from his upcoming book, TRACERS IN THE DARK). Hi-tech detective non-fiction at its finest. https://t.co/uk4XzqXQx9

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-22 16:43:49 This is touching (if overwrought), but it bothers me how it unintentionally valorizes an increasingly martial culture in the US, a belief that the right response to gun violence is for citizens to be armed and/or have military training https://t.co/xy0pz4H3mA

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-22 16:43:49 This is touching (if overwrought), but it bothers me how it unintentionally valorizes an increasingly martial culture in the US, a belief that the right response to gun violence is for citizens to be armed and/or have military training https://t.co/xy0pz4H3mA

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-22 16:43:49 This is touching (if overwrought), but it bothers me how it unintentionally valorizes an increasingly martial culture in the US, a belief that the right response to gun violence is for citizens to be armed and/or have military training https://t.co/xy0pz4H3mA

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-22 16:43:49 This is touching (if overwrought), but it bothers me how it unintentionally valorizes an increasingly martial culture in the US, a belief that the right response to gun violence is for citizens to be armed and/or have military training https://t.co/xy0pz4H3mA

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-22 16:43:49 This is touching (if overwrought), but it bothers me how it unintentionally valorizes an increasingly martial culture in the US, a belief that the right response to gun violence is for citizens to be armed and/or have military training https://t.co/xy0pz4H3mA

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw

2022-11-22 16:43:49 This is touching (if overwrought), but it bothers me how it unintentionally valorizes an increasingly martial culture in the US, a belief that the right response to gun violence is for citizens to be armed and/or have military training https://t.co/xy0pz4H3mA

2022-11-20 02:29:26 You know you’re in a San Francisco thrift store when https://t.co/vS4BZJZYzn

2022-11-18 16:34:13 I’m not saying Elon will succeed, but it certainly seems to be the approach he’s trying.

2022-11-18 16:32:40 So the story of Gideon in the Bible is of a visionary leader who took a large army, whittled it down to a tiny fraction through tests that weeded out all but the fanatically loyal, and led it to victory using completely unorthodox tactics designed to sow confusion in the enemy.

2022-11-18 06:26:27 @KarenAttiah Maybe the plan was to turn it into a dating site all along. $8 a month is cheaper than Tinder Gold and there's none of that annoying swiping

2022-11-18 06:23:05 @swodinsky WE did it. It's the glorious Twitter collective, the brave heroes of the socialmediaist international

2022-11-18 06:20:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And that’s why I think that, as immoral as it may seem to use an AI to use an artist’s creative labor for free and without consent, that is not the same as the question of how to protect an artist’s livelihood—which is the question that matters more. /

2022-11-18 06:19:28 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So what would the law look like if instead of restricting access to people’s data or IP, it focused on protecting people from the harms that can result from such access? Isn’t that the ultimate goal anyway? I’m curious how privacy and copyright experts are thinking about this.

2022-11-18 06:19:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam To me this makes the AI art question a subset of a broader question about the nature of our laws on data. Right now these laws are mostly designed around restricting ~access~ to data or IP. And they often fail to achieve even that.

2022-11-18 06:18:32 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But if he’s wrong, and AI art does take work away from artists, then excluding your art from the training data won’t shield you from that consequence. Protecting the creative labor from exploitation won’t protect its creators from losing work.

2022-11-18 06:16:21 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam And so the bigger question is whether gen-AI will rob artists of work, regardless of what data it’s trained on. Kevin’s piece is an argument that it won’t, and I think it’s an argument worth considering, which is why I published it.

2022-11-18 06:16:02 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam But let’s say any artist who wanted to could use such an amended copyright law to exclude their art from AI training data. There would still be a lot of imagery online to work from. Enough, I suspect, that gen-AI art would still be useful at least for many purposes.

2022-11-18 06:15:08 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam So perhaps we could amend copyright law such that AI art models can’t be trained on art unless it's public domain or has the right CC license. I’m sure you know more than me about how that could work.

2022-11-18 06:14:20 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam I think the real discomfort with genAI art though is that it’s a use of your creative labor you didn’t authorize. We’d need copyright law to recognize that someone might be using the fruits of your creative work even when they’re not directly reproducing the work.

2022-11-18 06:13:55 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam At one level, yes, using gen-AI art as training data is the same kind of immoral as ClearviewAI scraping your social media pictures or people analyzing trends using your public Venmo payments or Strava routes—it’s a use of your data you didn’t envisage.

2022-11-18 06:13:40 @FlyingTrilobite @neilturkewitz @kevin2kelly @WIRED @sciam Thanks for your patience. OK, here goes, while this site still exists....

2022-11-18 06:01:53 yay it worked https://t.co/VYPr3fVoRJ

2022-11-18 06:00:33 I'm https://t.co/QWCoAXGWfF But also, @elonmusk loves this site too much for it to disappear, he'll be at his laptop fixing bugs himself if that's what it takes to keep 116 million followers. So see you tomorrow

2022-11-18 05:57:35 QUICK FOLLOW JONATHAN IT'LL BE GOOD KARMA https://t.co/BalEmF8Twd

2022-11-18 05:05:29 "I find it a good philosophical exercise to imagine the last tweet." Maybe not so philosophical soon, @ftrain? https://t.co/iCCLQ9OwFI

2022-11-18 05:00:54 @GiladEdelman If newsroom productivity is having 27 mastodon instances open in tabs and trying to scan them all simultaneously then sure

2022-11-18 02:36:53 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED I have thoughts on this but they’re a little complex so bear with me while I find the time to turn them into tweets

2022-11-17 22:39:04 @neilturkewitz @FlyingTrilobite @kevin2kelly @WIRED This isn't saying artists are irrelevant. Without human art there'd be no AI-made art—obviously. All this says is that if an AI is trained on lots of art, then removing ~one~ artist's work from the training set won't make a big difference to what the AI produces.

2022-11-17 22:24:45 @GiladEdelman tried to warn you

2022-11-17 01:05:04 @quinnnorton He forgot to include lettuce

2022-11-16 23:51:01 RT @a_greenberg: My new book TRACERS IN THE DARK is currently the #1 True Crime audiobook on Amazon right now. Thank you to everyone for li…

2022-11-16 23:50:26 I can't tell if Jack's comment on this is a renunciation of responsibility, a shrug of despair, or a kind of Zen chant of acceptance I do know that you can't use bitcoin or any other crypto without trusting that a whole bunch of developers did their job right https://t.co/vZNcfCe1kQ https://t.co/pU0qpgiLNA

2022-11-16 23:35:39 It traces back to the insistence of Joel Kaplan, Facebook/Meta's policy chief, that conservative and liberal content be flagged to an equal degree, even if one side actually produced more misinformation than the other, as @BenWoffordDC reported for @WIRED. https://t.co/tlw7tOs7V4 https://t.co/iQbMXVRqzL

2022-11-16 23:34:52 Meta's policy of not fact-checking Trump as a presidential candidate (or any other politician) is indeed "long-standing," as a Meta spokesperson says here (1/2) https://t.co/ga3uAojAuD

2022-11-16 18:50:50 RT @stokel: Folks who have just 'd out of Twitter today: I want to talk to you. You lasted longer than others. You know more. Please email…

2022-11-16 18:50:31 @DKThomp @teddyschleifer @hunterwalk what'd i miss

2022-11-16 18:47:04 @6_Figure_Invest @reckless @Chiquita @dominos And of course there ~are~ serious technical challenges too, but they have to do with delivering the service at scale, which Twitter was basically doing fine

2022-11-16 18:45:30 @6_Figure_Invest @reckless @Chiquita @dominos Also, distinguish authentication from moderation. Authenticating or verifying who an account belongs to is a human task but a one-time thing. Moderating content is an ongoing problem and the tricky issues are policy/social, not technical—what is harassment, what is parody, etc.

2022-11-16 18:44:01 @6_Figure_Invest @reckless I think you're being overoptimistic about the sophisticated AI. A lot of corporate brands use social media very predictably, but look at, for instance, @Chiquita or @dominos. Quirky and unconventional approaches to social media are on the rise as brands try to stand out.

2022-11-16 17:52:59 Another example of thinking of Twitter as a technical engineering problem. It's not. It's a social-engineering problem (or as @reckless puts it, a content moderation service) https://t.co/zRwF1RnGsw